While HAART has significantly impacted the AIDS epidemic it will not end the epidemic; the need is to purge the body of infection which will require the use of drugs that reach into bone marrow, brain, any tissue that harbors latent HIV and clear or lower these viral reservoirs to make HAART unnecessary. Animal models were not required for design of antiretroviral drugs, but are an essential preclinical step to test the ability of drugs to induce latent HIV and eradicate latent viral reservoirs. Because of potential toxicity and unknown efficacy of such drugs, an effective animal model is critical for testing of therapeutic agents that will induce/eradicate latent HIV for design of clinical trials. We have a rigorous SIV model of HAART in HIVinfected individuals that reduces viral load in peripheral blood and cerebrospinal fluid (CSF) to undetectable levels and have latent virus in CD4+ T cells and macrophages in tissues, brain and spleen. HIV eradication requires HAART that includes CNS-penetrant antiretrovirals (ARTs) since brain is a site of viral latency. We will optimize a CNS-penetrant HAART regimen using current ART drugs using our SIV macaque, confirming that drugs reach effective levels in tissues quantitating CSF and brain levels of each antiretroviral drug. Using this SIV HAART model, the hypothesis that latent viral reservoirs (in addition to CD4+ T cells) in cells ofthe monocyte/macrophage lineage and potentially in CD34+ cells exist, and that an animal model is required to identify, quantitate and characterize these reservoirs. We hypothesize that viral infection can be eradicated from tissues by inducing virus reactivation from latently infected cells and preventing completion ofthe virus life cycle by ART that effectively penetrate tissues, including brain. We will characterize latent reservoirs in this SIV model in CD4+ T cells, monocytes, tissue macrophages and OD34+ cells (brain, spleen, bone marrow, lymph nodes, lung and gut). We will test induction/eradication drugs in SIV-infected macaques with HAART once the number of latently infected resting CD4+ T cells plateaus in plasma, quantitate residual virus replication and latently infected cells in plasma, CSF and spleen biopsies before and after induction therapy. We will examine the time to virus reactivation after withdrawal from HAART and successful induction/eradication therapy may result in no virus reactivation. We will use such a successful treatment regimen to dissect the molecular basis ofthis induction/eradication approach.